In Kidney Transplant Recipients (KTRs), non-invasive predictive biomarkers that allow early intervention are limited. Here we report findings from a prospective longitudinal study using B cell subsets and their cytokines as a biomarker in KTRs to predict rejection and outcome.

165 KTRs transplanted between dates 1/13-12/14 with serial biopsies (for-cause & protocol at 3 & 12mos) and blood drawn at 0,1,3,6 & 12mos served as the training set (TS). The results were validated (VS) in an additional prospective cohort of 74 KTRs transplanted between 1/15-12/15. IS was with Thymo induction, rapid steroid withdrawal, and TAC+MPA maintenance. Overall, 40% of KTRs had acute rejection (AR) by 1yr (24% subclinical; 16% clinical; all ACR & 5 ACR+ABMR). By 3 mos, 25% of patients had AR. 40% of these had persistent/recurrent AR at 1 yr despite initial therapy. Of the 75% KTRs with no rejection at 3 mos, 30% developed late AR (at 1yr). Neither clinical parameters (e.g. DSA, Creat) nor 3 mos histology could predict response to treatment or late AR.

We examined the cytokine profile (IL-10:TNFa expression ratio) in B cell subsets (by flowcytometry after CD40L & CpG stimulation) as a predictive biomarker for graft outcomes. At all time-points, KTRs with AR had a significantly lower IL-10:TNFα ratio within all B subsets, and especially within T1 transitional B cells (T1B). The T1B IL-10:TNFα ratio at 3 mos independently predicted AR in both training and validation sets (TS: ROC AUC 0.88, Sensitivity (sens) 87%, Specificity (spec) 86%, p<0.0001; VS: ROC AUC 0.97, sens 90% spec 94%, p<0.0001). Importantly, at 3 mos in patients with normal protocol biopsy, a low T1B IL-10:TNFα ratio (<1.26) predicted subsequent rejection (TS: AUC0.87, sens 88%, spec 81%, p<0.0001; VS: AUC0.98, sens 100%, spec 90%, p=0.04). Moreover, in KTRs treated for early AR, the T1B cytokine ratio at 3mos predicted response to therapy vs. persistent/recurrent AR (TS: ROC AUC 0.80, sens 78%, spec 82%, p=0.004; VS: AUC 78, sens 90%, spec 68%). Further, KTRs with a low T1B IL-10:TNFα ratio had significantly worse creatinine at 18 and 24 mos post-transplant and was independently associated with interstitial fibrosis with inflammation on 12 mos biopsy. Finally, this biomarker was not confounded by “inflammation” caused by infection (eg: BK/CMV viremia) or maintenance immunosuppression.

Conclusion: T1B IL-10:TNFα ratio at 3 mos was tested and validated as a strong and specific biomarker for subsequent renal allograft rejection, inflammation, and clinical course.

CITATION INFORMATION: Cherukuri A, Sharma A, Mehta R, Hariharan S, Rothstein D. Human Transitional B Cell Cytokine Profile Predicts Renal Allograft Rejection and Clinical Course. Am J Transplant. 2017;17 (suppl 3).

« Back to 2017 American Transplant Congress